Minor questions on how to run rebreather simulation scripts

Faceless

I've downloaded MatLab scripts from your site and succesfully run them.
Entire sequence of actions looks througly developed, pretty detailed and consequent. Alex, you and your team once again made a really great and complicate job here.

A minor questions on how to simulate failures during the dive.
I see pretty long list with most of possible breaks/falures, but don't understand how to add/remove them when simulation has started.
How can they be added? Do i need to put simulation on the pause for that ?

AD_ward9

Thanks for checking the script. As you are the first who seems to be asking the sort of questions one expects to arise when using a new program, we would like to ensure you get First Class service. If you send me an email with a phone number and time, we can walk you through and compile a Q&A of your questions to help those that follow behind you. Handy being in the same time zone as some of our developers, and same first language.

The short answer is you click on the fault you want to enable it and enter either the depth or the time in which they occur. Click again on them to remove them. I will write something up properly for Monday and post it here.

When some of our own staff started using the CCR model, they had a lot of questions: things that were unexpected, but because the model was in development at the time, we simply tried to fix it to make it more obvious, or gave them a verbal 1:1 answer. We now need to document use of the model better, and by logging carefully your queries as well as answering them properly, I hope we can achieve that.

Cheers,
Alex

Faceless

Well, cause I'm a huge fan and (hopefully) a potential owner of O.R. rebreather. i'm sure glad to be useful.
See PM with my email & phone.

Faceless

Here are first two questions "as for the beginning", they're mostly general.
1. Does setting failure modes has any impact on predive checks ?
I'v noticed that for example despite me setting "oxygen solenoid stuck open" mode or "diluent cylinder empty" failure mode options i'm always able to pass predive check.
2. Does the similation has any ability to plan dive like VPlanner or VR3? So far i've worked only with "profile mode" where dive simulation is being performed against existing dive profiles. How a new profile can be created in manual mode. Smth like me entering desired depth & time, next computer at predive stage checking validity of entered value and offering some profile to work with.

Once again i'm terribly sorry for probably silly questions, my experience with Rebreather so far is limited to several introductory dives, so i'm familiar with all the nesessary check & bailout procedures, but on a very limited basic. For my deco dives on open cycle i'm using VPlanner with "Best mix" calculator.

Faceless

A question regarding dive simulation in manual mode:
What is a list of actions to be performed to simulate descend/ascend to desired depth with some speed. For example: after i've hit "dive now" left button should i enter depth & descent speed before pressing "Run" button ?
Or it can be done only after setting simulation on Pause ?

AD_ward9

Bob Davidov will call you tomorrow and walk through it with you, as I am travelling in a few minutes time.

There are two ways of entering a profile:
1. As a set of depths and times to be at those depths,
2. As a set of descent rates and durations.

You can load a profile that you saved, or create one on the fly. To create one on the fly, hit run, and move from waypoint to waypoint: I usually do this by setting the depth to Pause at, then the descent rate.

Cheers,

Alex

Faceless

Here are first bugs:
I don't know maybe i'm doing smth wrong ? ...

When I ran dive simulation partially in manual mode, partially in deco, PPO2 some how has drop down to 0.40 !
Here is my dive profile:



And this is PPO2 graph


PPO2 setpoing was set to 1.3

Here is the link to the generated dive profile
http://orrb.faceless.ru/55m_deco_ac/55m_deco_ac.mat

Steps to reproduce:
1. Set gas mix to air ( 21%O2, 0% He) set descent speed to 30, and pause conditions to 55 meters. Start simulation.
2. When sim will go to pause after reaching 55 meters depth, set desc speed to 0, switch to accelerated mode, change pause conditions to wait for 20 minutes and press Run.
3. When system will calculate nesessary amount of steps and go to Pause again, change dive input to "deco" and hit Run button again.

AD_ward9

The PPO2 profile seems to be correct.

The reason it drops to 0.4 is because the initial descent caused the PPO2 to spike above 1.6. This means the diver's CNS clock was burnt up in the descent. The CNS clock lasts just seconds with PPO2 above 1.6.

So, with no CNS margin available anymore, the system reduces the PPO2 reduced automatically to 0.4 to avoid the diver having a convulsion.

The 1.6 limit is reduced if you are running the "Safe CNS" by a 0.2/0.3 margin. There is a thread on this, HERE.

The reason PPO2 rises at the end of the dive, is that if the diver is ascending and within 10m of the surface, it increases O2 to the same fraction of air, in case the diver decides to rocket to the surface. Unless this is done, the diver will die if he runs out of O2, or turned his O2 off near the surface: he would not have time to respond. This is one of the failure modes listed in the "How Rebreathers Kill People" list, and the solution is just what the simulation shows happens.

The spike of PPO2 at 25m is due to running into the CNS clock again: it has not wound down enough for the system to increase PPO2. You can see it trying to increase PPO2 again at 34 minutes, only to hit the same limit. Then at 42 minutes you are going for the surface, and are too close to the surface to allow the PPO2 to be lower than the fraction in air with a margin for your O2 consumption. It therefore raises PPO2 to avoid any risk of a hypoxic loop.

The final diagonal is the PPO2 reducing as you are on the surface.

So quite a useful profile, in demonstrating two of the hazards of rebreather diving: PPO2 spiking on descent burning up the CNS clock, giving rise to an O2 tox hazard if it is not dealt with, and the hazard of low PPO2 near the surface.

I missed one of your earlier questions, sorry, it was:
No, though the predive checks picks up many of them, the failures can occur during the dive. The purpose of the simulation is to check what the effect is of the failure on the ability of the system to keep the diver alive.

Our job is to find all routes by which you can kill the diver and prevent them, if at all possible. Some things like a breath hold from 50m are outside the scope, but anything attached to the Rebreather should be within scope.

Have some green for a good post!
Cheers,

Alex

AD_ward9

"Faceless" has written a nice "Dummy's Guide" to using the simulator.

Deserves some green, but the article seems to be a closed thread. It is HERE.

Thanks,

Bob Davidov reran the profile you posted and suggests you run in background in normal mode (not accelerated). Normal mode does updates once a second. This means the environment is handled much more accurately than just taking the way points, and joining the waypoints by straight lines.

Also, you have a double plot (two sets of lines). Bob corrected me: the last diagonal is joining your two plots, not simply returning to 0.21 on the surface (thanks for that: I did wonder why it went from right to left instead of the left to right I see when I do the same).

Bob's full explanation is attached as a small user advice note.

Alex

Attached Images

sim_advice.pdf (233.0 KB, 27 views)